Sociodemographic, clinical and organisational factors associated with delayed hospital discharges: a cross-sectional study.

Background Evidence from studies conducted in Western countries indicates that a significant proportion of hospital beds are occupied by patients who experience a delayed hospital discharge (DHD). However, evidence about this topic is lacking in Italy, and little is known on the patients’ and organisational characteristics that influence DHDs. Therefore, we carried out a survey in all the hospitals of a Northern Italian region to analyse the prevalence and the determinants of DHD. Methods A cross-sectional study was carried out during an index period of 15 days in 256 operative units in Emilia-Romagna, a Northern Italian region with 4.4 million inhabitants, to identify patients medically fit for discharge but still hospitalised. The characteristics of these patients (n = 510) were compared with all the other patients (n = 5,815) hospitalised in the same operative units during the index period using multilevel logistic regression models. Results The one-day prevalence of DHD was 8.1%. More than half of DHD patients (52.7%) waited to access long-term/rehabilitation units or residential care homes, 16.7% experienced a delay for family-related reasons, and 14.5% were waiting to be admitted to other rehabilitation services. Among DHD patients hospitalised in long-term/rehabilitation units, 45.3% were waiting to be transferred to residential care homes. Patients’ characteristics associated with a higher likelihood of DHD in multilevel logistic regression were older age, provision of intensive care, a diagnosis of dementia, tumours or femoral/shoulder fractures, and a number of comorbidities. Patients hospitalised in long-term/rehabilitation units, as well as in orthopaedics/traumatology units, were significantly more likely to have a DHD compared with patients hospitalised in general surgery units. Moreover, compared with Local Health Authority Hospitals, being hospitalised in Hospital Trusts was associated with a higher likelihood of DHD. Conclusions Although the prevalence of DHD in the present study is markedly lower than that reported in the literature, we submit that the DHD problem should be addressed with major organisational innovations, with a special focus on the ageing of the population and epidemiological trends. Organisational changes imply new ways of managing emerging clusters of patients whose needs are not efficiently or effectively met by traditional organisation models and services. Keywords: ; Patient flow; Survey; Intermediate care; Multilevel analysis